Google has just launched Gmail Mobile, a simplified XHTML interface for mobile phone with browsers. According to their Help Center:

Gmail Mobile lets you access your Gmail account from your mobile phone. Just enter http://m.gmail.com in your phone’s web browser to log in to your account.

Gmail Mobile offers a number of cool features:

• The interface is automatically optimized for the phone you’re using
• You can access attachments, including photos, Microsoft Word documents, and .pdf files
• If you enter phone numbers in your Gmail Contacts list, you can reply to messages by call

You can try it yourself by entering this URL: http://m.gmail.com

Does it mark an end to the gmail-lite/gmail-mobile projects? Not really.

• We will continue our maintenance of libgmailer, the PHP API for accessing Gmail, until, of course, they provide an official one, which I think is unlikely.
• Neerav’s gmail-mobile project sure won’t be stopped: its targets are WAP phones using WML, not HTML/XHTML.
• Some people might already get used to gmail-lite’s interface? Anyway, I believe Google folks would agree that it’s better for users to have choices.

Now, honestly, I am kind of excited. Google is now our "competitor"! That would be quite a challenge to make people to think that ours is even better than the official one, which is from Google!

posted by Gan, on December 16, 2005 under gmail-lite, gmail-mobile, libgmailer | Leave a Response

A bug of Gmail has been found which, if being abused, could give you access to any account. The bug was discovered in 14 Oct, by Anelkaos of elhacker.net. Anelkaos reported it to Google, which then fixed the bug 4 days later.

I have a very quick look at the hacking procedure. If I understand that correctly, it involves using an victim’s partial "authentication string" - something like session ID or the long, seemingly random string sometimes you could see in URLs - to fake Gmail into believing that you were logged in as the victim. The procedure looks complicated, but can definitely be eased by a script or something.

Anyhow, it is glad that Google has fixed it rather quickily. 

posted by Gan, on November 19, 2005 under gmail-lite, Web Tech | 1 Comment

I hosted this site on my own server before moving it to sourceforge. At that time, I used a free, lite version of WebLog Expert to generate site traffic information from the Apache log files. It was good.

Of course this method no longer works as I do not have access to the log files now. I did not track the traffic, until Google announced Google Analytics, a free traffic-analyzing service for everyone.

Oh, Google again! Those skeptics said, they are more and more like Microsoft, who keep acquiring and expanding and using "advanced" technology until there is no room (and money) for its competitors.  

But according to Eric Peterson, a senior analyst on web site technology, Google does not seem to be forcing other analytics vendors out of business. There are still many websites, most of them blogs, that didn’t heard of or use any website analysis services until Google started providing a free one. The market still has plenty of room for everyone at the moment.

Actually I do not worry too much about Google being the Big Brother, as I think the chance is slim. But it is in general not a good idea to put all eggs in one basket. Given the fact that I have used too many Google’s products, this time I would really like to try software from other companies. However, most of them, for example those mentioned by Eric, are either not free or by invitation (e.g. Measure Map) only. If any one knows a good substitution, please let me know.

Before I found another solution, I would keep using Google Analytics. Nonetheless, I think my visitors here should have the right to know this: This site uses Google Analytics to track YOU. 

posted by Gan, on November 19, 2005 under Google | Leave a Response

Mozilla have released Firefox 1.5 RC1 for interested parties to download. According to their roadmap, the stable version will be released within this year.

I would love to try this RC out. I didn’t, however, because I did not know which extensions that I have installed would be broken.

The release candidate is supposed to help extension developers to make their extensions "in sync." Unfortunately, some of them are developed by amateur programmers and, for whatever reason, they no longer give support nor update to their extensions. These extensions can be quite useful and some users simply can’t live without them. For example, the Tong Wen Tang （同文堂）extension is very handy to me as well as a lot of Chinese users, but the author has stopped updating it since Firefox 0.9. Luckily a little hack can make it compatible with Firefox 1.0+. But how about 1.5?

Extension is the second most important reason why I switched to Firefox (security is no. 1). Ironically, they are so useful that I might decide not to update my Firefox if I can’t keep using them in the new version! I believe that other people, especially those who have installed a lot of extensions, may share the same feeling.

posted by Gan, on November 6, 2005 under Web Tech | 1 Comment

Wired is one of my favourite magazine. I enjoy reading every single page of it, including advertisements. Yes, I know I can read most of the articles in its website. But it is those little corners or boxes that really enlightens me. For example, in this issue (November 2005) I have learnt that:

•  LifeGem is a company that would turn your remains into a diamond that lasts (almost) forever. "The company filters out the calcium, turns your carbon into graphite, and puts you in an oven for six months. With lots of heat and pressure, the carbon atoms realign into a diamond."
• Think you have the best home videos and want to share them with any one visiting your… grave? Try Vidstone’s Serenity Panel! "Utilizing solar-power technology and a weatherproof LCD panel it provides families the option of viewing a personalized video tribute right at their loved one’s final resting place. It features a 5-10 minute multimedia memorial detailing the most precious memories of your loved one’s life." Okay, seriously, can I play p0rn?
• "Maps, guides, and other reference works include bits of bogus information - fake streets, phony restaurants - so the original publishers can finger competitors who lift their work." But they can’t sue them for copyright infringement, however, since "courts have rules that, just like actual facts, faux facts can’t be copyrighted." I wonder if I could sue the publisher if the misinformation leads me to a damage.

Finally, don’t miss out the artifacts from the future!

posted by Gan, on November 3, 2005 under Others | Leave a Response

Ajax is hot. Some people even claimed that web applications, or web apps, based on this technology will one day dominate the market. Many people retold Sun’s motto some years ago: "Network is the computer." And suddenly, Google, with its rich "lab" of web apps, becomes the most threatening enemy of Microsoft.

For security reason, browsers, and hence web apps, cannot access to your local files. Some say this is a critical limitation of web apps. But who needs local files if most of your apps are web-based? Actually, web apps nowadays normally allow you to open/save files located in their own domain.

However, cross-domain file access is usually not supported. For example, the photos you saved in Flickr cannot be accessed from Writely. Well, Flickr is popular, so some web apps, mostly blog systems, do allow you to read or write from it. But this is in a rather ad hoc fashion. Why isn’t there a "web filesystem" that other web apps can access to easily?

Google is a company (if not the only company) that has the ability to provide such service globally. In fact, I am writing this blog because I have the feelings that Google Base is actually heading to this direction. Imagine there is an API such that data can be import/export from the Base in the format of, say, XML through the HTTPS protocol. Then other web apps can use it to read/write "files" to this universal "filesystem"! Moreover, the API would allow web apps to extract metadata instead of just the file itself. This is even more useful, I think! Not to mention the ability to search, efficiently.

Of course, security is a big issue that must be examined very, very carefully, because you are authorizing whatever web app to access to a pool of data owned by you. But this doesn’t stop me from getting excited about this very idea!

Let me make a bolder statement: the main battle of the war between Microsoft and Google is actually WinFS vs. Google Base! Microsoft, inherently favors local applications, invests heavily in this next generation, database-driven local filesystem; while Google, being the most powerful web company in the World, puts its bet on the data generated by rich web applications. Woo! I would really like to see the strategies they will use to win this war!

posted by Gan, on November 2, 2005 under Web Tech, Google | Leave a Response

Google have launched their "universal database" service, called Google Base. The service is semi-open now. I am yet to be able to access it, but according to people that can use it, this is what Google promise:

Google Base is Google’s database into which you can add all types of content. We’ll host your content and make it searchable online for free.

Examples of items you can find in Google Base:

• Description of your party planning service
• Articles on current events from your website
• Listing of your used car for sale
• Database of protein structures

You can describe any item you post with attributes, which will help people find it when they search Google Base. In fact, based on the relevance of your items, they may also be included in the main Google search index and other Google products like Froogle and Google Local.

This is clever. Google aims to make all information in the world best organized. Who else better than the owner of data to tell Google what kind of data they are having, and exactly how to organize it?

And the name of it… I think everyone would agree that All Your Base Are Belong To Google…

More screenshots can be found here.

I wish they would release an API for this. If they don’t, maybe we can implement ours! libgooglebase! Woo! 

posted by Gan, on October 31, 2005 under Google | 1 Comment

Gmail has really changed its name to "Google Mail" officially - but only to the users in the UK.

Even the logo has to be changed:
vs.

In addition, all new signups from the UK will be assigned "@googlemail.com" addresses. Although existing accounts will be able to use either domain for now.

To most users, including me, the scariest thing that could happen is that we were all forced to change our currently using email address from your.name@gmail.com to your.name@googlemail.com. In fact, according to their help page (emphasis mine):

We are still working with the courts and trademark office to protect our ability to use the Gmail name, but in the meantime, we want you to have an email address you can rely on.

So Google is suggesting that, starting from today, not to give other people your email address as your.name@gmail.com, since it can be invalid at any time in the future. Use your.name@googlemail.com instead. It is a valid address even for user outside the UK (you can try it yourself). In fact, you may treat it as an alias to your @gmail.com address at this moment. All mails sent to your.name@googlemail.com will appear in your.name@gmail.com mailbox.

 

Even though I am not a UK Gmail user, and I seldom post my Gmail address to the public, I really hope I do not have to forgo my @gmail.com address: my.name@googlemail.com sounds sooooo unimpressive!

posted by Gan, on October 21, 2005 under Google | 7 Comments

Because of a new law suit, Gmail might not be able to keep using the name Gmail.

This is an old news, and Google have already done something to “unload” the name. The changes have been done quietly that the general public may not notice, but not to us developer. For example, because of its change in URL from gmail.google.com/gmail to mail.google.com/mail, gmail-lite and libgmailer failed to use immediately. Fortunately we discovered that and fixed it quickily. And now it seems that you cannot find any references to the word “Gmail” anywhere in its AJAX source code.

Google would probably rename Gmail back to “Google Mail”. Boring name, but safe. After this, I think Google will no longer “invent” new names for its new products. They will keep on using the pattern “Google xxxx”: Google Maps, Google Reader, Google Suggests, Google Desktop, Google OS (okay I faked this one), etc etc. You can see them in Google Labs.

What about all the “hacks” based on it, such as gmail-lite? Should we change to googlemail-lite and libgooglemailer also?

I think I will keep on using this name, unless they, uh, shoot sue me.

posted by Gan, on October 14, 2005 under Google | 1 Comment

Gmail has further extended its support in other languages. Now you can check your spelling in, not just English, but also French, German, Italian, Dutch, etc.

Unfortunately, no “word checking” for Chinese at this moment.

posted by Gan, on October 14, 2005 under Google | Leave a Response